Wang et al. Cancer Cell Int (2018) 18:64 https://doi.org/10.1186/s12935-018-0563-6 Cancer Cell International

PRIMARY RESEARCH Open Access MicroRNA‑365 promotes lung carcinogenesis by downregulating the USP33/ SLIT2/ROBO1 signalling pathway Yuhuan Wang1, Shuhua Zhang1, Hejing Bao2, Shukun Mu1, Baishen Zhang1, Hao Ma3 and Shudong Ma1*

Abstract Background: Abnormal microRNA expression is closely related to cancer occurrence and development. miR-365a-3p plays an oncogenic role in skin cancer, but its role in lung cancer remains unclear. In this study, we aimed to investi- gate its role and underlying molecular mechanisms in lung cancer. Methods: Western blot and real-time quantitative PCR (qPCR) were used to detect the expression of miR-365a-3p in lung adenocarcinoma and lung cancer cell lines. The efects of miR-365a-3p on lung cancer cell proliferation, migra- tion, and invasion were also explored in vitro. The potential miR-365a-3p that targets USP33 was determined by dual luciferase reporter assay and verifed by qPCR and western blot analysis. miR-365a-3p acts as an oncogene by promot- ing lung carcinogenesis via the downregulation of the miR-365a/USP33/SLIT2/ROBO1 axis based on western blot analysis. Subcutaneous tumourigenesis further demonstrated that miR-365a-3p promotes tumour formation in vivo. Results: miR-365a-3p was upregulated in lung adenocarcinoma and lung cancer cell lines. Overexpression of miR- 365a-3p promoted and inhibition of miR-365a-3p suppressed the proliferation, migration, and invasion of lung cancer cells. We identifed USP33 as the downstream target of miR-365a-3p and observed a negative correlation between miR-365a-3p and USP33 expression in lung adenocarcinoma patients. The miR-365/USP33/SLIT2/ROBO1 axis, a new mechanism, was reported to inhibit the invasion and metastasis of lung cancer. A nude mouse model of lung cancer further verifed these fndings. Conclusions: In summary, miR-365a-3p acts as an oncogene by promoting lung carcinogenesis via the downregula- tion of the USP33/SLIT2/ROBO1 signalling pathway, making the miR-365/USP33/SLIT2/ROBO1 axis a new mechanism of lung cancer promotion and a novel therapeutic target for predicting prognosis and response to therapy. Keywords: Adenocarcinoma of lung, Carcinogenesis, Deubiquitinating enzymes, MicroRNAs, Oncogenes

Background medicine [3–5]. Terefore, identifying that drive According to the World Health Organization (WHO), cancer progression and providing efective treatments cancer is the leading cause of death worldwide, and lung may signifcantly prolong the survival of patients with cancer is the main cause of cancer-related deaths [1]. In non-small cell lung cancer (NSCLC) [6, 7]. 2017, there were an estimated 222,500 new diagnoses of MicroRNAs (miRNAs) are small, highly conserved, lung cancer in the United States, and 155,870 people died noncoding RNAs of approximately 19–25 nucleotides of lung cancer [2]. With advances in medical technology, that regulate at the post-transcriptional lung cancer treatment has entered the era of precision level by base-pairing with the 3′ UTRs of target mRNAs [8–11]. Tey play key roles in human biological processes, such as migration, cellular metabolism, cell proliferation, *Correspondence: [email protected] 1 Department of Oncology, Nanfang Hospital, Southern Medical apoptosis, and epithelial–mesenchymal transition (EMT) University, Guangzhou 510515, Guangdong, China [12–17]. Increasing evidence has demonstrated that miR- Full list of author information is available at the end of the article NAs play an important role in cancer and are closely

© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat​iveco​mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat​iveco​mmons​.org/ publi​cdoma​in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wang et al. Cancer Cell Int (2018) 18:64 Page 2 of 14

related to tumourigenesis and prognosis. Te miRNA signalling participates in the development of cancer [29, miR-365 gene is located on 16p13.12, and 30]. Tus, in this study, we aimed to elucidate the role the mature hsa-miR-365 sequence is cleaved from two of miR-365a-3p in lung cancer cells and the relationship precursors: hsa-miR-365-1 and hsa-miR-365-2 [18]. among miR-365a-3p, USP33, SLIT2, and ROBO1 in lung Abnormal expression of miR-365 is observed in a variety adenocarcinoma. of tumours, with diferent expression patterns and func- tions in diferent human cancer types. In skin squamous Methods cell carcinoma, we frst found that miR-365a-3p plays Tissue samples an oncogenic role by downregulating NFIB to promote Tis study was approved by the Ethics Review Board of CDK6 and CDK4 expression, leading to Rb phospho- Nanfang Hospital, Southern Medical University (Guang- rylation and tumour progression [18–20]. Additionally, zhou, China). Twenty pairs of primary lung cancer tis- high expression levels of miR-365a-3p can be detected in sues and their corresponding adjacent non-tumour tissue breast and pancreatic cancer, playing a role in promot- samples were collected in the Toracic Surgery Depart- ing tumour development [21, 22]. In colon cancer, miR- ment of Nanfang Hospital from March to June 2017. 365a-3p inhibits the development of cancer by targeting All experiments were performed according to relevant cyclin D1 and BCL-2 [23, 24]. Tis miRNA also serves guidelines; informed consent was obtained from each as a tumour suppressor gene in gastric cancer [25]. In patient. No patients received prior radiotherapy or chem- brief, the functions of miR-365a-3p in cancer are com- otherapy. All specimens were removed during surgery plex. Tus, the roles of miR-365a-3p in lung cancer cells and immediately stored at − 80 °C in liquid nitrogen for remain unclear and require further study. subsequent extraction of total RNA. Deubiquitinating enzymes are key enzymes that can reverse ubiquitination modifcations. Recently, their Cell lines and cell culture functions and mechanisms have been of great interest in A549 cells were purchased from the American Type Cul- tumour research [26, 27]. USP33, also known as VHL- ture Collection (ATCC, Manassas, VA, USA). SPC-A-1 interacting 1 (VDU1), is located and H1299 cells were purchased from the Shanghai Cell on and encodes two subtypes: Bank of the Chinese Academy of Sciences (Shanghai, type I and type II. Type I contains 942 amino acids, and China). All cells were cultured in RPMI-1640 medium or type II contains 911 amino acids, with predicted molecu- DMEM supplemented with 10% FBS (Gibco, Gaithers- lar weights of 107 and 103 kDa, respectively. USP33 has burg, MD, USA). All cells were maintained in a humidi- been confrmed to inhibit the development of a variety of fed incubator at 37 °C and 5% CO­ 2. tumour cells. USP33 deubiquitinates and thus maintains the stability of ROBO1, thereby regulating the activity RNA extraction and real‑time quantitative RT‑PCR of SLIT2 and inhibiting cancer cell metastasis [28–30]. (qRT‑PCR) USP33 maintains the stability of CP110 and antagonises Total RNA was extracted from cell lines or tissues using the efect of cyclin-F on the S/G2 and G2/M phases of a TRIzol Kit (Takara Bio, Shiga, Japan) according to the the cell cycle by interacting with CP110, maintaining the manufacturer’s instructions. cDNA was synthesized using steady state of the centriole and ensuring normal func- Takara RT reagent (Takara Bio). qRT-PCR was performed tioning of mitosis and genome stability, which reduces on a Light Cycler 480 system (Roche Diagnostics, Basel, tumourigenesis [31]. Switzerland) using a SYBR Green I Master kit (Roche). Additionally, SLIT2/ROBO1 signalling has been shown We used glyceraldehyde-3-phosphate dehydrogenase to inhibit tumour cell proliferation and migration. Prasad (GAPDH) as an internal reference. Mature miR-365a-3p et al. [32] found that ROBO1 and ROBO2 were expressed expression was measured by qRT-PCR according to the in several breast cancer cell lines and that SLIT2 inhib- Taqman MicroRNA Assay protocol (Takara Bio) and ited CXCL12/CXCR4-induced chemotaxis, invasion, normalised using U6 small nuclear RNA with the 2­ −ΔΔCt adhesion, and secretion of MMP-2 and MMP-9 in breast method. cancer cells. Te loss of SLIT2, SLIT3, or ROBO1 protein in mouse breast cancer models results in an increase in Western blotting repair processes in tissues, promoting cell proliferation Equal amounts of protein were separated by 10% SDS- [33]. PAGE and blotted onto PVDF membranes (Millipore, Our previous studies showed that in cutaneous squa- Bedford, MA, USA) probed with the following primary mous cell carcinoma, miR-365a-3p plays an oncogenic and secondary antibodies: monoclonal rabbit primary role by promoting tumour progression [18]. Moreover, antibodies against SLIT2 and ROBLO1 (1:1000; Cell studies have shown that USP33-mediated SLIT2/ROBO1 Signaling Technology, Boston, MA, USA) and β-tubulin Wang et al. Cancer Cell Int (2018) 18:64 Page 3 of 14

(1:10,000; Bioworld Technology Inc., St. Louis Park, MN, precoated with Matrigel. Each assay was performed at USA); polyclonal rabbit primary antibody against USP33 least three times independently. (1:500; Abcam, San Francisco, CA, USA); and secondary fuorescent goat anti-rabbit antibody (LI-COR, Lincoln, Wound healing assay NE, USA). Primary antibodies were applied overnight at When cells had grown to approximately 90% confuency 4 °C, and secondary antibody treatment was performed (after 48 h), an artifcial wound was created with a 10-µL for about 1 h at 25 °C. An Odyssey Infrared Imaging pipette tip. Cells were then cultured in fresh medium System (LI-COR) was used to analyse immunoreactive without FBS. Images were taken at 0 and 36 h to visual- bands. Western blotting was performed three times. ise wound healing. Te relative percentage of the wound healed was calculated using the following formula: (width Plate clone formation assay of wound at 0 h − width of wound at 36 h)/width of A549 or SPC-A-1 cells were seeded into a 6-well culture wound at 0 h. plate (200 cells/well) and incubated for 12 days. Cells were stained with Giemsa solution. Plates were scored Plasmid and oligonucleotide construction by determining the number of colonies containing ≥ 50 AntagomiR-365, antagomiR-negative control (antag- cells. omiR-NC), agomiR-365, and agomiR-negative con- trol (agomiR-NC) were designed and synthesized by 5‑Ethynyl‑2ʹ‑deoxyuridine (EdU) assay GenePharma (Shanghai, China). pcDNA3.1-USP33 and 5-Ethynyl-2ʹ-deoxyuridine incorporation assays were pcDNA3.1 vectors were designed and synthesized by conducted using the EdU assay kit (RiboBio Co., Guang- Obio Technology (Shanghai, China). zhou, China) according to the manufacturer’s instruc- tions. Cells were incubated with 50 nM EdU for 2 h at Transient transfection 37 °C. Cells were then fxed with 4% formaldehyde for A549 and SPC-A-1 cells were seeded in 6-well plates at a 15 min at 25 °C and treated with 0.5% Triton X-100 for density of 30–50%. Transient transfection was performed 20 min at 25 °C to permeate cell membranes. After wash- with Lipofectamine 2000 reagents (Invitrogen, Carlsbad, ing with PBS three times, cells were incubated with 1× CA, USA) according to the manufacturer’s instructions. Apollo reaction cocktail (100 µL/well) for 30 min. DNA For all experiments, cells were collected 24–28 h after was stained with 10 µg/mL of Hoechst 33342 stain transfection. (100 µL/well) for 20 min, and staining was visualised with fuorescence microscopy. Five felds of view were ran- Dual‑luciferase reporter assay domly selected for each sample. EdU-positive cells were Te dual-luciferase reporter plasmids psi-CHECK2- stained with red dye, and the relative proliferation-posi- USP33 (containing the wild-type USP33 3ʹ UTR bind- tive ratios were calculated from the average cell count of ing site) and psiCHECK2-mGPC3 (containing a mutant the fve visualised felds. USP33 3ʹ UTR) were constructed. A549 or SPC-A-1 cells were added to 24-well plates at 70–80% confuence 24 h Cell migration and invasion assays before transfection. A mixture of 50 nM miR-365a-3p Te migratory and invasive abilities of cells were assessed agomiR or antagomiR and 0.5 µg psi-CHECK2 reporter using Transwell inserts (Corning, Inc., Corning, NY, plasmid (psiCHECK2-wUSP33 or psiCHECK2-mUSP33) USA) in 24-well plates. For invasion assays, each group was co-transfected into cells using Lipofectamine 2000 4 of cells (5 × 10 cells/100 µL) was resuspended in FBS- reagent. At 48 h after transfection, luciferase activity was free RPMI-1640 medium and seeded into the upper analysed using a dual-luciferase reporter assay system chamber containing a Matrigel-coated membrane. After (Promega, Madison, WI, USA) according to the manu- incubation for 24 h at 37 °C with 5% CO­ 2, the incuba- facturer’s instructions. Each experiment was performed tion medium and non-invading cells were removed from in triplicate. the upper surface of the membrane with cotton swabs. Invading cells that adhered to the lower surface of the In vivo tumourigenesis assays chamber were fxed in 4% paraformaldehyde for 20 min All animal experimental protocols were approved by the and stained with 0.1% crystal violet for 30 min. Invading Animal Research Ethics Committee of Nanfang Hospital cells were photographed and manually counted at 200× and complied with the rules of the specifc pathogen-free magnifcation using a microscope (Olympus, Tokyo, (SPF) animal laboratory of the Nanfang Medical Univer- Japan). For the Transwell migration assays, the process sity. 4- to 6-week-old male mice were purchased from the was the same, except the Transwell membrane was not Animal Laboratory Center of Nanfang Medical Univer- sity (Guangdong, China). A549/antagomiR-365 cells and Wang et al. Cancer Cell Int (2018) 18:64 Page 4 of 14

Fig. 1 miR-365a-3p is upregulated in non-small cell lung cancer tissues and cell lines. a Relative expression levels of miR-365a-3p were analysed by qRT-PCR in 20 paired human lung adenocarcinoma tissues and adjacent matched normal paracancerous tissues (n 20 per group). b Relative = expression levels of miR-365a-3p were analysed by qRT-PCR in the normal lung epithelial cell line 16HBE and lung cancer cell lines A549, SPC-A-1, and H1299. Results were normalised with U6 in each sample. Experiments were repeated at least three times independently. Data represent mean SD. ***P < 0.001 ±

A549/antagomiR-NC cells were injected subcutaneously Statistical analyses were performed using SPSS 22.0 soft- 6 into the left and right axilla of fve nude mice (4 × 10 ware (SPSS, Chicago, IL, USA). Signifcant diferences cells on each side) beginning 8 days after the injection were analysed using Student’s t tests for continuous vari- of tumour cells. Te mice were sacrifced 26 days post- ables. Spearman’s correlation was used to analyse the injection to observe changes in tumour volume. Tumour relationship between miR-365a-3p and USP33 mRNA size was measured every 3 days using the same protocol, expression; P < 0.05 indicated signifcance. and tumour volumes were calculated with the following 2 3 Results formula: V = (L × W )/2, where V is the volume ­(mm ), L is the biggest diameter (mm), and W is the smallest diam- miR‑365a‑3p is upregulated in NSCLC tissues and cell lines eter (mm). To evaluate the expression of miR-365a-3p in lung ade- nocarcinoma cells, qRT-PCR assays were performed Bioinformatics analysis in 20 patients with lung adenocarcinoma and matched Te Cancer Genome Atlas (TCGA) data portal provides paracancerous tissues. U6 was used as an internal con- a platform for researchers to query, download, and ana- trol. Results showed that the average expression level of lyse data sets generated by TCGA (http://cance​rgeno​ miR-365a-3p in cancer tissues was higher than that in me.nih.gov/). the corresponding paracancerous tissues (fold difer- Bioinformatics analysis performed with miRanda/Tar- ence = 2.560, P = 0.0329; Fig. 1a). Next, the expression getScan/starBase indicated that the 3′ UTR of USP33 is a levels of miR-365a-3p were analysed in various lung can- binding site for miR-365a-3p. cer cell lines. qRT-PCR results demonstrated that higher expression levels of miR-365a-3p were present in A549, Statistical analysis SPC-A-1, H1299, and PC9 lung cancer cell lines than Data are expressed as mean ± SD. Each experiment was in 16HBE normal lung epithelial cells. Tis was con- repeated at least three times unless otherwise indicated. sistent with our clinical fndings, indicating increased

(See fgure on next page.) Fig. 2 miR-365a-3p promotes the proliferation, migration, and invasion ability of lung cancer cells. a Plate clone formation efciencies of A549 or SPC-A-1 cells following overexpression of miR-365a-3p by agomiR or inhibition of miR-365a-3p by antagomiR treatment, as compared with controls. b Representative images of A549 or SPC-A-1 cells stained with Hoechst (blue) and EdU (red), which indicates DNA synthesis, after transfection with agomiR-365, antagomiR-365, or controls. c Wound healing assay in A549 or SPC-A-1 cells transfected with agomiR-365, antagomiR-365, or controls. d Transwell migration and invasion assay in A549 or SPC-A-1 cells transfected with agomiR-365, antagomiR-365, or controls. For all experiments, agomiR-NC and antagomiR-NC were used as the negative controls for agomiR-365 and antagomir-365, respectively. Left panel, representative images. Right panel, quantitative data. All experiments were performed in triplicate and repeated three times independently. *P < 0.05, **P < 0.01, ***P < 0.001 compared to controls Wang et al. Cancer Cell Int (2018) 18:64 Page 5 of 14 Wang et al. Cancer Cell Int (2018) 18:64 Page 6 of 14

miR-365a-3p expression in lung adenocarcinoma tis- USP33 is a downstream target gene of miR‑365a‑3p sues (Fig. 1b). Additionally, among the lung cancer cell and is downregulated by miR‑365a‑3p lines, miR-365a-3p exhibited the highest expression in Te above in vitro experiments demonstrated that miR- A549 cells and the lowest expression in SPC-A-1 cells; 365a-3p is closely correlated with the proliferation, therefore, we selected these two cell lines for subsequent migration, and invasion of lung cancer cells. Analy- experiments. Taken together, the above results indicated sis using multiple databases showed that USP33 may that the abnormal expression of miR-365a-3p may be be a target gene of miR-365a-3p. TCGA data showed a involved in lung carcinogenesis. negative correlation between miR-365a-3p and USP33 expression, and survival analysis demonstrated that high miR‑365a‑3p promotes proliferation, migration, expression of USP33 was associated with good progno- and invasion in lung cancer cells sis (Additional fle 2). Additionally, it was previously Our fndings suggested that the dysregulated expres- reported that USP33 exhibits low expression in lung ade- sion of miR-365a-3p may result in lung cancer progres- nocarcinoma tissues and cells [30, 34]. Tus, we sought to sion. To explore the function of miR-365a-3p, A549 determine whether miR-365a-3p interacts with USP33. A and SPC-A-1 cells were transfected with agomiR-365 western blot assay showed that the expression of USP33 or antagomiR-365 (Additional fle 1). To investigate the was signifcantly downregulated following the overex- efect of high expression of miR-365a-3p on the prolifera- pression of miR-365a-3p (Fig. 3a). Next, qRT-PCR was tion of lung cancer cells, we used plate clone formation used to detect the expression of miR-365a-3p and USP33 and EdU assays. Te results of the plate clone formation in tissues derived from 20 cases of lung adenocarcinoma. assay showed that the overexpression of miR-365a-3p Consistent with our hypothesis, we observed a nega- increased the number of colonies (P < 0.05), whereas the tive correlation between the expression of miR-365a-3p 2 opposite result was observed following the inhibition of and that of USP33 in these tissues (R = 0.2467, P < 0.05; miR-365a-3p expression (P < 0.05; Fig. 2a). Te EdU assay Fig. 3b). To further verify that miR-365a-3p directly tar- results revealed that cellular fuorescence intensity was gets and downregulates USP33, the wild-type 3′ UTR of increased after transfection with agomiR-365 (P < 0.05) USP33 (WT USP33 3′ UTR) or a mutated USP33 3′ UTR and decreased after transfection with antagomiR-365 (Mut USP33 3′ UTR) was cloned into a dual-luciferase (P < 0.05; Fig. 2b). Tus, the EdU assay results were con- UTR vector and then co-transfected with agomiR-365 sistent with those of the plate clone formation assay, or a negative control (agomiR-NC) into A549 and SPC- further confrming that miR-365a-3p promotes the pro- A-1 cells. At 48 h after transfection, cells were harvested liferation of lung cancer cells. and lysed to detect luciferase activity (Additional fle 3, Next, we performed wound healing and Transwell Additional fle 4: Table S1). Te results showed that rela- migration and invasion assays. Te wound healing assay tive luciferase activity was signifcantly decreased in cells demonstrated that wound healing rates were elevated co-transfected with WT USP33 3′ UTR and agomiR-365. following transfection with agomiR-365 (P < 0.05) and However, there was no signifcant change in luciferase reduced following transfection with antagomiR-365 activity after co-transfection with Mut USP33 3′ UTR (P < 0.05; Fig. 2c). In the Transwell migration and invasion and agomiR-365 (Fig. 3c). Tis suggested that miR- assay, the numbers of cells that migrated and invaded the 365a-3p directly targets and binds the USP33 3′ UTR. Matrigel were elevated following transfection with ago- miR-365 (P < 0.05) and reduced following transfection miR‑365a‑3p promotes lung cancer cell proliferation, with antagomiR-365 (P < 0.05; Fig. 2d). Tus, the results migration, and invasion by downregulating USP33 of the Transwell migration and invasion assay were con- Te above experiments demonstrated that miR-365a-3p sistent with those of the wound healing assay, indicating targets USP33 and downregulates its expression. Next, that miR-365a-3p promotes the migration and invasion functional rescue assays were performed to further ver- activities of lung cancer cells. ify that the miR-365a-3p-mediated downregulation of

(See fgure on next page.) Fig. 3 USP33 is a downstream target gene of miR-365a-3p and is downregulated by miR-365. a Expression levels of USP33 were detected by western blotting of A549 or SPC-A-1 cells after transfection with agomiR-365, antagomiR-365, or controls. β-tubulin was used as an internal control. Experiments were performed at least three times, and representative bands are shown. b (a) Sequence alignments of miR-365a-3p binding sites in the wild-type and mutant USP33 3ʹ UTRs. The replaced site is underlined. (b) Wild-type USP33 3ʹ UTR or mutant USP33 3ʹ UTR was co-transfected with agomiR-365 or agomiR-NC into A549 or SPC-A-1 cells. After 48 h, relative luciferase activities were measured. *P < 0.05, **P < 0.01, ***P < 0.001. c Expression levels of miR-365a-3p and USP33 were examined by qRT-PCR in 20 paired human lung adenocarcinoma tissues and adjacent normal lung tissues. A correlation was observed between miR-365a-3p and USP33 expression as determined by Spearman correlation analysis (n 20; = R2 0.2467, P < 0.05) = Wang et al. Cancer Cell Int (2018) 18:64 Page 7 of 14 Wang et al. Cancer Cell Int (2018) 18:64 Page 8 of 14 Wang et al. Cancer Cell Int (2018) 18:64 Page 9 of 14

(See fgure on previous page.) Fig. 4 miR-365a-3p promotes lung cancer cell proliferation, migration, and invasion by downregulating USP33. a, b The proliferative ability of A549 or SPC-A-1 cells after overexpression of USP33 by transfection of pcDNA3.1-USP33, miR-365a-3p by transfection with agomiR, or USP33 and miR-365a-3p simultaneously was measured using the a plate clone formation and b EdU assays. pcDNA3.1-vector and agomiR-NC were used as negative controls. c, d The migration and invasion activities of A549 or SPC-A-1 cells transfected with pcDNA3.1-USP33, agomiR-365, or both vs. controls were measured using the c Transwell migration and invasion and d wound healing assays. Left panel, representative images. Right panel, quantitative data. All experiments were performed at least in triplicate and repeated three times. *P < 0.05, **P < 0.01, ***P < 0.001 compared to controls

USP33 promotes the proliferation, migration, and inva- ROBO1 were upregulated following the overexpression sion of lung cancer cells. Te pcDNA3.1-USP33 plasmid of USP33 (Fig. 5a, b) but downregulated following the was constructed to overexpress USP33, with the empty overexpression of miR-365a-3p (Fig. 5c). However, when pcDNA3.1 vector used as a negative control (Additional both miR-365a-3p and USP33 were overexpressed, there fle 1b). Te plate clone formation assay results showed was no signifcant change in the expression of SLIT2 or that the numbers of cell colonies derived from A549 and ROBO1 compared to levels in the control group (Fig. 5d). SPC-A-1 cells co-transfected with pcDNA3.1-USP33 and Tese results indicated that miR-365a-3p promotes lung agomiR-NC were signifcantly reduced (P < 0.05), whereas cancer progression by downregulating the USP33/SLIT2/ these numbers increased markedly compared with those ROBO1 signalling pathway. in negative controls after co-transfection with pcDNA3.1 vector and agomiR-365 (P < 0.05). In contrast, the simul- miR‑365a‑3p promotes subcutaneous tumour formation taneous overexpression of USP33 and miR-365a-3p by in mice co-transfection with pcDNA3.1-USP33 and agomiR-365 Te results of this in vivo experiment showed that resulted in no signifcant changes in the number of cell tumourigenic ability began to difer on day 8 post-injec- colonies compared to that in the control group (Fig. 4a). tion. Beginning on day 8, tumour size changes were Te EdU assay results revealed that the overexpression of measured every 3 days using a slide rule, and tumour USP33 in combination with the overexpression of miR- growth curves were plotted (Fig. 6). Te tumour volumes 365a-3p mitigated an increase in the cellular fuorescence on day 26 in mice injected with A549/antagomiR-NC intensity caused by the overexpression of miR-365a-3p cells were signifcantly larger than those in mice injected alone, consistent with the results of the plate clone for- with A549/antagomiR-365. Our fndings demonstrated mation assay (Fig. 4b). Tis further suggested that miR- that tumours derived from A549/antagomiR-365 cells 365a-3p promotes the proliferation of lung cancer cells grew more slowly than those derived from A549/antag- by downregulating USP33. Similarly, the wound healing omiR-NC cells. Tus, the results of the subcutaneous assay showed that the wound healing rate increased in implantation tumour model further demonstrated that cells overexpressing miR-365a-3p alone and decreased miR-365a-3p promotes tumour formation in vivo. in the presence of pcDNA3.1-USP33 (Fig. 4c). Te same patterns were observed in the Transwell migration and Discussion invasion assay (Fig. 4d). Identifying the mechanisms of lung cancer metastasis and methods for inhibiting metastasis is an important miR‑365a‑3p regulates the USP33/SLIT2/ROBO1 signalling focus of the international medical community. MiRNAs pathway play a dual role in tumour development and progression Te above series of in vitro experiments suggested that and participate in all aspects of tumour development miR-365a-3p promotes lung cancer progression by [35–38]. Circulating miR-365a-3p may serve as a molec- downregulating USP33; however, the signalling path- ular marker for the diagnosis and prognostic evaluation ways involved in this process remain unknown. It has of some tumours [39–41]. Tus, the role of miR-365a-3p been reported that the USP33/SLIT2/ROBO1 signalling in cancer should be further explored. pathway inhibits the migration of cancer cells; there- Te regulation of miR-365a-3p difers among diferent fore, we hypothesised that miR-365a-3p downregu- tissues and genetic backgrounds, in which it can serve lates this pathway to promote lung cancer progression. as either a tumour suppressor or oncogene [42, 43]. Te To verify our hypothesis, a western blot assay was per- dual nature of miR-365a-3p may be related to its unique formed to observe changes in the expression of SLIT2 topological structure, interactions with various signal- and ROBO1 in A549 and SPC-A-1 cells overexpressing ling pathways, and induction of diferent biochemi- USP33 or miR-365a-3p alone or in combination. Te cal reactions. It is therefore necessary to study the role results showed that the expression levels of SLIT2 and of miR-365a-3p in specifc tumour contexts. Although Wang et al. Cancer Cell Int (2018) 18:64 Page 10 of 14 Wang et al. Cancer Cell Int (2018) 18:64 Page 11 of 14

(See fgure on previous page.) Fig. 5 miR-365a-3p regulates the USP33/SLIT2/ROBO1 signalling pathway. Western blot was used to measured SLIT2 and ROBO1 expression levels following overexpression of a, b USP33, c miR-365, and d both USP33 and miR-365a-3p simultaneously in SPC-A-1 and A549 cells. Each experiment was performed at least three times, and representative bands are shown

lung cancer is the most common cause of cancer-related Ubiquitination is a dynamic and reversible process, deaths worldwide, the role of miR-365a-3p in the prolif- and the removal of ubiquitination modifcations is eration, invasion, and metastasis of lung cancer has not mainly mediated by deubiquitinases, which are closely been previously reported. As lung adenocarcinoma is related to tumour occurrence and development [44–46]. the main pathological type of lung cancer, it is important In recent years, increasing attention has been paid to the to study the role and regulation of miR-365a-3p in lung role of deubiquitinases in tumour progression, and the adenocarcinoma. molecular mechanisms underlying their regulation have In this study, we found that miR-365a-3p was highly gradually been clarifed [47]. More than 40 deubiquit- upregulated in cancer tissues and cell lines. Based on inating enzymes have been found to be associated with these experimental results, we speculated that miR- the occurrence and development of tumours. However, 365a-3p may promote tumourigenesis in lung cancer. few studies have investigated the relationship between Functional assays demonstrated that miR-365a-3p pro- miRNAs and tumourigenesis in the context of deubiq- moted proliferation, migration, and invasion, further uitination. USP33 is a deubiquitinating enzyme that confrming our hypothesis. However, the pathway by is closely related to the occurrence and development which miR-365a-3p promotes lung cancer remained of tumours [28–30, 32]. Te list of diseases currently unknown. Tus, determining the role and mechanism of reported to be associated with the dysregulation of miR-365a-3p in lung adenocarcinoma was the frst aim of USP33 includes breast cancer, acute lymphoblastic leu- this study. kaemia, and lung cancer [28, 30, 48]. We sought to iden- tify the downstream target genes of miR-365a-3p and

Fig. 6 miR-365a-3p promotes subcutaneous tumour formation in mice. a Images show subcutaneous tumour growth at 26 days after the injection of A549 cells transfected with antagomiR-365 or control into the axilla of fve BALB/c nude mouse xenograft models. Beginning 8 days after injection, tumour volumes ­(mm3) were measured every 3 days, and b time-dependent tumour growth curves were plotted. Data are expressed as mean SD ± Wang et al. Cancer Cell Int (2018) 18:64 Page 12 of 14

found that USP33 expression was strongly negatively Additional fles correlated with that of miR-365a-3p in lung adenocar- cinoma tissues. Functional rescue assays demonstrated Additional fle 1. Transfection efciency of miR-365a-3p and USP33. (a) that miR-365a-3p promoted proliferation, migration, Transfection efciency of agomiR-365 and antagomiR-365 in A549 and and invasion by downregulating USP33. Tis is the SPC-A-1 cells. (b) Transfection efciency of pcDNA3.1-USP33 in A549 and SPC-A-1 cells. frst evidence for a lung cancer invasion and metastasis Additional fle 2. Association of USP33 expression with survival. (a) Sur- mechanism in which miR-365a-3p directly targets and vival analysis showing the association between high expression of USP33 downregulates USP33, which further promotes the pro- and good prognosis. liferation, migration, and invasion activities of lung can- Additional fle 3. Construction of luciferase vector and identifcation of cer. However, these results did not clarify the signalling mutant USP33 3ʹ UTR by PCR. (a) USP33 3ʹ UTR PCR amplifcation product. pathway through which miR-365a-3p downregulates (b) Positive clones identifed by colony PCR. (c) psi-CHECK2 vector. USP33 to promote lung carcinogenesis. Additional fle 4: Table S1. PCR sequences of miR-365a-3p and USP33. (a) hsa-miR-365a-3p sequence. (b) USP33 3 UTR primer sequences. Recently, an increasing number of studies have found ʹ that the SLIT2/ROBO1 signalling pathway is closely involved in tumourigenesis by inhibiting the prolifera- Abbreviations tion, migration, and invasion of tumour cells [49–51]. miRNAs: microRNAs; NSCLC: non-small cell lung cancer; EMT: epithelial– mesenchymal transition; VDU1: VHL-interacting deubiquitinating enzyme 1; USP33 deubiquitinates ROBO1 in lung cancer cells to GAPDH: glyceraldehyde-3-phosphate dehydrogenase; SPF: specifc pathogen- maintain its stability, thereby regulating SLIT2 activity free; TCGA:​ The Cancer Genome Atlas. and inhibiting cancer cell metastasis [30]. Based on the Authors’ contributions above experiments, we hypothesised that miR-365a-3p SDM guided the entire experiment. YHW performed the experiments and may regulate the SLIT2/ROBO1 signalling pathway by produced the fgures. YHW and SHZ wrote the manuscript. HJB and SKM targeting USP33 to promote lung cancer. Based on west- designed the study. BSZ analysed the data. HM was involved in collabora- tions regarding data acquisition. All authors have contributed signifcantly to ern blot assay results, the expression levels of SLIT2 and the content of the manuscript. All authors read and approved the fnal ROBO1 were both downregulated following the over- manuscript. expression of miR-365a-3p but restored when USP33 Author details was overexpressed at the same time. Combined with 1 Department of Oncology, Nanfang Hospital, Southern Medical University, the above in vitro test results, we concluded that miR- Guangzhou 510515, Guangdong, China. 2 Department of Oncology, Chong- 365a-3p promotes lung cancer by downregulating the qing Three Gorges Center Hospital, Chongqing, China. 3 Department of Clini- cal Medicine, Tianjin Medical University College, Tianjin, China. USP33/SLIT2/ROBO1 pathway. To further validate this hypothesis, we performed Acknowledgements in vivo experiments in a mouse lung cancer tumour The authors are grateful to all staf at the study centre who contributed to this study. model. A549/antagomiR-365 cells were injected into mice to observe subcutaneous tumour formation. Te Competing interests results suggested that the inhibition of miR-365a-3p The authors declare that they have no competing interests. expression reduced tumourigenicity. Tus, the in vivo Availability of data and materials experiment further demonstrated that miR-365a-3p All data generated or analysed during this study are included in this published plays a role in promoting tumour proliferation in lung article [and its additional fles]. adenocarcinoma. Consent for publication This manuscript is approved by all authors for publication in Cancer Cell Conclusions International.

In summary, we not only confrmed that miR-365a-3p Ethics approval and consent to participate targets and downregulates USP33 in lung adenocarci- This study was approved by the Ethics Review Board of Nanfang Hospital, noma, but we also further confrmed that miR-365a-3p Southern Medical University (Guangzhou, China). All experiments were performed according to relevant guidelines; informed consent was obtained promotes the proliferation, migration, and invasion of from each patient. All animal experimental protocols were approved by the lung cancer cells by downregulating the USP33/SLIT2/ Animal Research Ethics Committee of Nanfang Hospital and complied with ROBO1 pathway. To achieve early diagnosis and fur- the rules of the specifc pathogen-free (SPF) animal laboratory of the Nanfang Medical University. ther improve the clinical outcomes of lung cancer, it is necessary to explore biomarkers for early detection and Funding develop individualised treatment methods for lung can- This study was supported by Grants from the Natural Science Foundation of Xinjiang Uygur Autonomous Region, China (2017D01C004). cer. Our elucidation of the mechanism by which miR- 365a-3p mediates the USP33/SLIT2/ROBO1 signalling Publisher’s Note pathway in lung carcinogenesis thus provides a new strat- Springer Nature remains neutral with regard to jurisdictional claims in pub- egy for the diagnosis and targeted therapy of lung cancer. lished maps and institutional afliations. Wang et al. Cancer Cell Int (2018) 18:64 Page 13 of 14

Received: 3 March 2018 Accepted: 21 April 2018 25. Guo SL, Ye H, Teng Y, Wang YL, Yang G, Li XB, et al. Akt-p53-miR-365-cyclin D1/cdc25A axis contributes to gastric tumorigenesis induced by PTEN defciency. Nat Commun. 2013;4:2544. 26. Fraile JM, Quesada V, Rodríguez D, Freije JM, López-Otín C. Deubiqui- tinases in cancer: new functions and therapeutic options. Oncogene. References 2012;31:2373–88. 1. World Health Organization (WHO). GLOBOCAN 2012: estimated cancer 27. Suresh B, Lee J, Kim KS, Ramakrishna S. The importance of ubiquitina- incidence, mortality and prevalence worldwide in 2012. 2018. http:// tion and deubiquitination in cellular reprogramming. Stem Cells Int. globo​can.iarc.fr/Defau​lt.aspx. Accessed 13 Jan 2018. 2016;2016:6705927. 2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 28. Yuasa-Kawada J, Kinoshita-Kawada M, Rao Y, Wu JY. Deubiquitinating 2017;67:7–30. enzyme USP33/VDU1 is required for signaling in inhibiting breast 3. Collins FS, Varmus H. A new initiative on precision medicine. N Engl J cancer cell migration. Proc Natl Acad Sci USA. 2009;106:14530–5. Med. 2015;372:793–5. 29. Huang Z, Wen P, Kong R, Cheng H, Zhang B, Quan C, et al. USP33 medi- 4. Borczuk AC, Allen TC. PD-L1 and lung cancer: the era of precision-ish ates Slit-Robo signaling in inhibiting colorectal cancer cell migration. Int J medicine? Arch Pathol Lab Med. 2016;140:351–4. Cancer. 2015;136:1792–802. 5. Ashley EA. The precision medicine initiative: a new national efort. JAMA. 30. Wen P, Kong R, Liu J, Zhu L, Chen X, Li X, et al. USP33, a new player in lung 2015;313:2119–20. cancer, mediates Slit-Robo signaling. Protein Cell. 2014;5:704–13. 6. Chen SC, Lin MC, Chang JW, Wang SW, Lee CH, Tsao TC. Phase II study of 31. Li J, D’Angiolella V, Seeley ES, Kim S, Kobayashi T, Fu W, et al. USP33 regimen of gemcitabine and cisplatin in advanced non-small cell lung regulates centrosome biogenesis via deubiquitination of the centriolar cancer. Jpn J Clin Oncol. 2000;30:494–8. protein CP110. Nature. 2013;495:255–9. 7. Kris MG, Johnson BE, Berry LD, Kwiatkowski DJ, Iafrate AJ, Wistuba II, et al. 32. Prasad A, Fernandis AZ, Rao Y, Ganju RK. Slit protein-mediated inhibition Using multiplexed assays of oncogenic drivers in lung cancers to select of CXCR4-induced chemotactic and chemoinvasive signaling pathways targeted drugs. JAMA. 2014;311:1998–2006. in breast cancer cells. J Biol Chem. 2004;279:9115–24. 8. Lagos-Quintana M, Rauhut R, Lendeckel W, Tuschl T. Identifcation of 33. Marlow R, Strickland P, Lee JS, Wu X, PeBenito M, Binnewies M, et al. SLITs novel genes coding for small expressed RNAs. Science. 2001;294:853–8. suppress tumor growth in vivo by silencing Sdf1/Cxcr4 within breast 9. Cho WC. OncomiRs: the discovery and progress of microRNAs in cancers. epithelium. Cancer Res. 2008;68:7819–27. Mol Cancer. 2007;6:60. 34. Simicek M, Lievens S, Laga M, Guzenko D, Aushev VN, Kalev P, et al. 10. Lynam-Lennon N, Maher SG, Reynolds JV. The roles of microRNA in The deubiquitylase USP33 discriminates between RALB functions in cancer and apoptosis. Biol Rev. 2009;84:55–71. autophagy and innate immune response. Nat Cell Biol. 2013;15:1220–30. 11. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. 35. Bhaskaran M, Wang Y, Zhang H, Weng T, Baviskar P, Guo Y, et al. Cell. 2004;116:281–97. MicroRNA-127 modulates fetal lung development. Physiol Genomics. 12. Poy MN, Eliasson L, Krutzfeldt J, Kuwajima S, Ma X, MacDonald PE, et al. 2009;37:268–78. A pancreatic islet-specifc microRNA regulates insulin secretion. Nature. 36. Lizé M, Herr C, Klimke A, Bals R, Dobbelstein M. MicroRNA-449a levels 2004;432:226–30. increase by several orders of magnitude during mucociliary diferentia- 13. Brennecke J, Hipfner DR, Stark A, Russell RB, Cohen SM. bantam tion of airway epithelia. Cell Cycle. 2010;9:4579–83. encodes a developmentally regulated microRNA that controls cell 37. Kumar MS, Erkeland SJ, Pester RE. Suppression of non-small cell lung proliferation and regulates the proapoptotic gene hid in Drosophila. Cell. tumor development by the let-7 microRNA family. Proc Natl Acad Sci 2003;113:25–36. USA. 2008;105:3903–8. 14. Ebert MS, Sharp PA. Roles for microRNAs in conferring robustness to 38. Xu Z, Xiao SB, Xu P. miR-365, a novel negative regulator of interleukin-6 biological processes. Cell. 2012;149:515–24. gene expression, is cooperatively regulated by Sp1 and NF-kappaB. J Biol 15. Hamada S, Satoh K, Fujibuchi W, Hirota M, Kanno A, Unno J, et al. MiR-126 Chem. 2011;286:21401–12. acts as a tumor suppressor in pancreatic cancer cells via the regulation of 39. Sozzi G, Pastorino U, Croce CM. MicroRNAs and lung cancer: from markers ADAM9. Mol Cancer Res. 2012;10:3–10. to targets. Cell Cycle. 2011;10:2045–6. 16. Hamada S, Satoh K, Miura S, Hirota M, Kanno A, Masamune A, et al. miR- 40. Yu SL, Chen HY, Chang GC, Chen CY, Chen HW, Singh S, et al. MicroRNA 197 induces epithelial–mesenchymal transition in pancreatic cancer cells signature predicts survival and relapse in lung cancer. Cancer Cell. by targeting p120 catenin. J Cell Physiol. 2013;228:1255–63. 2008;13:48–57. 17. Singh SK, Pal Bhadra M, Girschick HJ, Bhadra U. MicroRNAs–micro in size 41. Vannini I, Fanini F, Fabbri M. MicroRNAs as lung cancer biomarkers and but macro in function. FEBS J. 2008;275:4929–44. key players in lung carcinogenesis. Clin Biochem. 2013;46:918–25. 18. Zhou M, Liu W, Ma S. A novel onco-miR-365 induces cutaneous squa- 42. Li M, Liu L, Zang W, Wang Y, Du Y, Chen X, et al. miR-365 overexpression mous cell carcinoma. Carcinogenesis. 2013;34:1653–9. promotes cell proliferation and invasion by targeting ADAMTS-1 in breast 19. Zhou L, Wang Y, Ou C, Lin Z, Wang J, Liu H, et al. microRNA-365-targeted cancer. Int J Oncol. 2015;47:296–302. nuclear factor I/B transcriptionally represses cyclin-dependent kinase 6 43. Bai J, Zhang Z, Li X, Liu H. MicroRNA-365 inhibits growth, invasion and and 4 to inhibit the progression of cutaneous squamous cell carcinoma. metastasis of malignant melanoma by targeting NRP1 expression. Int J Int J Biochem Cell Biol. 2015;65:182–91. Clin Exp Pathol. 2015;8:4913–22. 20. Zhou M, Zhou L, Zheng L. miR-365 promotes cutaneous squamous cell 44. Fraile JM, Manchado E, Lujambio A, Quesada V, Campos-Iglesias D, Webb carcinoma (CSCC) through targeting nuclear factor I/B (NFIB). PLoS ONE. TR, et al. USP39 deubiquitinase is essential for KRAS oncogene-driven 2014;9:e100620. cancer. J Biol Chem. 2017;292:4164–75. 21. Singh R, Saini N. Downregulation of BCL2 by miRNAs augments drug- 45. Fraile JM, Campos-Iglesias D, Rodríguez F, Español Y, Freije JM. The deu- induced apoptosis—a combined computational and experimental biquitinase USP54 is overexpressed in colorectal cancer stem cells and approach. J Cell Sci. 2012;125:1568–78. promotes intestinal tumorigenesis. Oncotarget. 2016;7:74427–34. 22. Hamada S, Masamune A, Miura S, Satoh K, Shimosegawa T. MiR-365 46. Wu Y, Wang Y, Yang XH, Kang T, Zhao Y, Wang C, et al. The deubiquitinase induces gemcitabine resistance in pancreatic cancer cells by targeting USP28 stabilizes LSD1 and confers stem-cell-like traits to breast cancer the adaptor protein SHC1 and pro-apoptotic regulator BAX. Cell Signal. cells. Cell Rep. 2013;5:224–36. 2014;26:179–85. 47. He M, Zhou Z, Shah AA, Zou H, Tao J, Chen Q, et al. The emerging role of 23. Nie J, Liu L, Zheng W, Chen L, Wu X, Xu Y, et al. microRNA-365, down- deubiquitinating enzymes in genomic integrity, diseases, and therapeu- regulated in colon cancer, inhibits cell cycle progression and promotes tics. Cell Biosci. 2016;6:62. apoptosis of colon cancer cells by probably targeting Cyclin D1 and Bcl-2. 48. De Pittà C, Tombolan L, Dell’Orto MC, Accordi B, te Kronnie G, Romualdi Carcinogenesis. 2012;33:220–5. C, et al. A leukemia-enriched cDNA microarray platform identifes new 24. Wang XY, Wu MH, Liu F, Li Y, Li N, Li GY, et al. Diferential miRNA expres- transcripts with relevance to the biology of pediatric acute lymphoblastic sion and their target genes between NGX6-positive and negative colon leukemia. Haematologica. 2005;90:890–8. cancer cells. Mol Cell Biochem. 2010;345:283–90. Wang et al. Cancer Cell Int (2018) 18:64 Page 14 of 14

49. Werbowetski-Ogilvie TE, Seyed Sadr M, Jabado N, Angers-Loustau A, 51. Mertsch S, Schmitz N, Jeibmann A, Geng JG, Paulus W, Senner V. Slit2 Agar NYR, Wu J, et al. Inhibition of medulloblastoma cell invasion by Slit. involvement in glioma cell migration is mediated by Robo1 receptor. J Oncogene. 2006;25:5103–12. Neurooncol. 2008;87:1–7. 50. Wang LJ, Zhao Y, Han B, Ma YG, Zhang J, Yang DM, et al. Targeting Slit- Roundabout signaling inhibits tumor angiogenesis in chemical-induced squamous cell carcinogenesis. Cancer Sci. 2008;99:510–7.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions